dbt Semantic Layer

dbt Semantic Layer centralizes metric definitions on dbt models and makes them available to analytics tools, applications, and AI systems. It is best suited to data teams that already work in dbt and need governed metrics shared across several destinations. Its code-based approach is compelling, but access requires a qualifying paid dbt account.

Overview

The Semantic Layer addresses a familiar analytics problem: the same business metric can be defined differently in different tools. Teams define metrics once on dbt models, and MetricFlow generates SQL and handles joins when those metrics are queried. That makes it a strong fit for organizations seeking consistent definitions across BI, embedded analytics, and AI, especially when dbt is already central to their data work.

This is a cloud-deployed product with API and web access. It supports Snowflake, BigQuery, Databricks, and Redshift. The main constraint is account eligibility: defining and querying metrics requires a dbt Starter or Enterprise-tier account. Single-tenant accounts also need setup and enablement from an account representative.

Key features

Semantic models and metrics are defined in YAML as version-controlled, lineage-aware code. This gives data teams a reviewable home for metric logic and can reduce the drift that comes from maintaining definitions separately in each consumer tool. The benefit is greatest when teams are prepared to manage metrics as part of their dbt development workflow.

MetricFlow converts those configurations into SQL at query time, handles joins, and rewrites queries to reduce compute and latency. It also optimizes joins and caches repeated queries. These mechanisms target efficiency as well as consistency; they do not remove the need to consider the workload against the monthly queried-metric allowance.

GraphQL and JDBC APIs, along with a Python SDK, provide ways to query metrics and dimensions. The product also offers an API, the dbt MCP Server, and out-of-the-box integrations for BI, notebooks, spreadsheets, embedded analytics, and AI systems. Claude and ChatGPT can connect to governed metrics through MCP, making this a relevant option for teams extending metric access to AI tools rather than leaving those tools to reconstruct business logic.

The dbt Labs Trust Center carries SOC 2 Type II, ISO 27001:2022, ISO 27017, ISO 27018, ISO 42001, GDPR, and CCPA badges. Starter includes 24x5 support; Enterprise offers priority support with SLAs.

Pricing

The Developer plan is free and includes one developer seat, one project, and 3,000 successful models per month. It does not include the Semantic Layer, so it is not a free way to use this product.

Starter costs $100 a month, billed per user/month. It includes five developer seats, one project, 15,000 successful models per month, and 5,000 queried metrics per month, as well as the Semantic Layer. This is the entry point for teams that need the product, but the per-user charge and metric cap matter as adoption grows.

Enterprise uses custom pricing and includes a custom developer seat count, 30 projects, 100,000 successful models per month, and 20,000 queried metrics per month. Enterprise+ also has custom pricing and the same model and metric allowances, but includes unlimited projects. These tiers suit larger deployments that need more projects or higher allowances than Starter provides; Enterprise+ is the clearer fit when project count cannot be kept within a fixed allocation.

A free trial is offered. Review the per-user Starter cost and queried-metric allowance against the expected number of consumers before committing.

Platforms

The Semantic Layer is available through web and API platforms and is deployed in the cloud. Its documented warehouse support covers Snowflake, BigQuery, Databricks, and Redshift.

Who it's for

Choose dbt Semantic Layer when a data team wants one governed, code-defined set of metrics for BI, applications, and AI access, and is willing to operate within dbt's account tiers and monthly query limits. It is less suitable for teams seeking a standalone free semantic layer: the free Developer plan does not include the product, and usage requires Starter or Enterprise-tier access.

Pros and cons

• Pro: YAML definitions are version-controlled and lineage-aware, giving teams a shared, reviewable basis for metric logic.
• Pro: MetricFlow handles SQL generation and joins, while APIs, SDK access, integrations, and MCP extend metrics across analytics and AI destinations.
• Pro: Starter includes the Semantic Layer and 24x5 support; Enterprise adds priority support with SLAs.
• Con: The free Developer plan excludes the Semantic Layer, so there is no free tier for using this capability.
• Con: Starter limits queried metrics to 5,000 per month and charges per user, which can constrain or raise the cost of broader adoption.
• Con: Single-tenant accounts require enablement by an account representative before use.

Alternatives

Cube is worth considering for teams that want API, Linux, self-hosted, and web options and a free tier capped at five workbooks, 1,000 daily requests, and one-day query-history retention.

Starburst Enterprise Context Layer may suit teams looking for a free option with up to three clusters and standard execution mode for ad hoc queries, plus a free trial.

Sema offers a free Sandbox with one connector, one user, 100 chat queries per month, and five saved reports, making it an alternative for a narrowly scoped chat-query setup.

Strata has a free plan with one developer, 25 users, two data sources per project, and unlimited projects and branches; consider it when those allowances match the team's needs.

Definite offers a free plan for two users and two connectors with 1 GB of storage, five credits per month, and daily sync.

Malloy is a free alternative available across API, extension, desktop, self-hosted, and web platforms.

Verdonz has a free guided workspace with sample data but no own business data, so it is an option for exploring a workspace rather than bringing business data into the free plan.

MetricFlow is free under the Apache 2.0 license and requires a working dbt project and dbt adapter; it is the alternative to consider for teams seeking the metric engine itself rather than dbt Semantic Layer's hosted product access.

For broader comparisons, see Semantic Layer Software, Database Documentation Tools, and Data Transformation Tools.

Verdict

dbt Semantic Layer is a strong choice for dbt-centered data teams that need governed metrics shared across analytics, applications, and AI. Its defining advantage is version-controlled metric logic paired with managed query access; look elsewhere if free use is essential or if the monthly allowances and dbt account requirement do not fit your deployment.